EXCEED_TOKEN_QUOTA_LIMIT

Encountering an error? Get help with AI tools

Error Cause

The large model Token usage has exceeded the quota limit. Different providers have different quota policies. You need to identify your current provider to determine the specific cause.

Identify Your Provider

Check the first argument of createModel("xxx") in your code — the argument value is your current provider. For example:

const model = ai.createModel("cloudbase");   // Provider is cloudbase
const model = ai.createModel("hunyuan-v3");  // Provider is hunyuan-v3

The provider name also appears in the request URL path:

https://cloud1-xxx.api.tcloudbasegateway.com/v1/ai/{provider}/chat/completions

The {provider} in the path is your current provider.

⚠️ Note: Text and image generation use different providers

• Text models (chat, content generation, etc.): Use createModel("cloudbase") or createModel("hunyuan-v3")
• Image models (image generation, etc.): Use createImageModel("hunyuan-image")

Choose the correct provider and API based on your actual usage.

Provider Quota Details

Provider	Status	Quota Details	Solution
cloudbase	✅ Supported	Uses free quota first; automatically falls back to plan quota when free quota is exhausted (requires resource point plan)	Both free and plan quotas are exhausted. Go to the CloudBase Console to purchase a resource pack or upgrade your plan
hunyuan-v3	✅ Supported	Only uses free quota; does not support plan quota	Free quota is exhausted. Switch to the cloudbase provider and purchase a resource point plan
hunyuan-exp	❌ Discontinued	No free quota available, unusable	Switch to a resource point plan and change the provider to cloudbase
deepseek	❌ Discontinued	No free quota available, unusable	Switch to a resource point plan and change the provider to cloudbase
Custom provider	—	CloudBase does not limit Token usage for custom models	Contact your third-party model provider for quota details

💡 Difference between cloudbase and hunyuan-v3

• cloudbase: Supports both free and plan quota consumption. When free quota runs out, it automatically uses plan quota. Suitable for users who have purchased a resource point plan. Model switches (e.g., hy3, hy3-preview) need to be enabled manually in the console.
• hunyuan-v3: Only supports free quota consumption. Reports an error when free quota is exhausted, with no plan quota fallback. Suitable for users without a resource point plan who only use free quota. No manual model switch configuration required.

For details, see: AI Mini Program Growth Plan - Provider Comparison

Solution

1. Identify your provider: Check the createModel() argument in your code.
2. Check the table above: Determine your provider's status and quota details.
3. Take the appropriate action:
   • cloudbase users → Purchase a resource pack or upgrade your plan to increase your quota.
   • hunyuan-v3 users → Switch your provider to cloudbase and purchase a resource point plan for plan quota support.
   • hunyuan-exp / deepseek users → Switch to a resource point plan and change the provider to cloudbase.
   • Custom provider users → Contact your third-party model provider for quota and billing information.